AIME 1995: Artificial Intelligence in Medicine pp 103-114 | Cite as
Semi-qualitative models and simulation for biomedical applications
Abstract
Simulation of biomedical systems (e.g. drug metabolism) is an important component of decision support in many medical applications. To cope with the uncertainty of the parameters that are used in the models, one may rely on either applying stochastic methods to quantitative simulation or using qualitative simulation instead, where a number of techniques infer as much as possible in the lack of quantitative information. Many cases, however, rather than lack of quantitative information, are cases of incomplete information, in that numerical bounds can be assigned to the unprecisely known parameters of a model. In such cases, constraint solving techniques might be used to cope with such source of uncertainty and a balance must be sought between the expressive power of the models and the constraint solving capabilities that can be effective applied to these models. This paper presents semi-qualitative modeling and simulation, a new approach aimed at reaching such appropriate balance. This new formalism is introduced with a simple example and by means of a more formal presentation, and its application to a multicompartmental model for drug metabolism is discussed.
Keywords
Quantitative Information Negative Rate Expressive Power Threshold Point Qualitative StatePreview
Unable to display preview. Download preview PDF.
References
- 1.R. Hofestadt, Petri Nets to Model Metabolic Processes, in Proceedings of MIE'94, pp. 115–120, Lisbon, May 1994.Google Scholar
- 2.E. Ackerman, L. Gatewood, J. Rosevear and G. Mollnar, Blood Glucose Regulation and Diabetes, in Concepts and Models in Biomathematics, Marcel Dekker, New York-Basel, 1969.Google Scholar
- 3.U.G Oppel, A. Hierle, L. Janke and W. Moser, Transformation of Compartmental Models into Sequences of Causal Probabilistic Networks, in Proceedings of AIME'93, IOS Press, pp. 75–83, Munich, October 1993.Google Scholar
- 4.B. Kuipers, Commonsense Reasoning about Causality: Deriving Behavior from Structure, Artificial Intelligence 24, pp.169–203, 1984.CrossRefGoogle Scholar
- 5.K.D. Forbus, Qualitative Process Theory, Artificial Intelligence 24, pp. 85–168, 1984.CrossRefGoogle Scholar
- 6.B. Kuipers and J. Kassirer, Causal Reasoning in Medicine: Analysis of a Protocol, Cognitive Science 8, pp. 363–385, 1984.CrossRefGoogle Scholar
- 7.L. Ironi, A. Catanneo and M. Stefanelli, A tool for pathophysiological knowledge acquisition, in Proccedings of AIME'93, IOS Press, pp. 13–31, Munich, October 1993.Google Scholar
- 8.B. D'Ambrosio, Qualitative Process Theory Using Linguistic Variables, Springer-Verlag, 1989.Google Scholar
- 9.B. Kuipers and D. Berleant, Using Incomplete Quantitative Knowledge into Qualitative Simulation, Technical Report AI90-122, Artificial Intelligence Laboratory, Dept. Of Computer Sciences, University of Texas at Austin, 1990.Google Scholar
- 10.B. Kuipers and D. Berleant, Combined Qualitative and Numerical Simulation with Q3, in Recent Advances in Qualitative Physics (B. Faltings, P. Struss eds.), MIT Press, 1992.Google Scholar
- 11.P. Barahona, A Causal and Temporal Reasoning Model and its Use in Drug Therapy Applications, Artifificial Intelligence in Medicine vol. 6, no. 1, pp. 1–28, 1994.CrossRefGoogle Scholar
- 12.J. Jaffar and S. Michaylov, Methodology and Implementation of a CLP System, in Proceedings Fourth International Conference of Logic Programming, Melbourne, 1987.Google Scholar
- 13.F. Azevedo, P. Barahona and F. Ferreira da Silva, Modeling causal and temporal knowledge to support therapy planning, in Proceedings of MIE'94, pp.109–114, Lisbon, May 1994.Google Scholar